Neuroblastoma is the most common extra-cranial solid tumor of childhood with an appalling 30% cure rate in children with advanced disease. There is a clear need for new chemotherapeutics, as current drugs are only marginally effective at the high doses that result in toxic acute and grave long term side effects. The overall objective of this proposal is to employ modern techniques of protein engineering to develop a new generation of non-immunogenic and pharmacologically optimized enzymes for chemotherapy of neuroblastomas and other central nervous system (CNS) cancers through L-Methionine (L-Met) depletion. L-Met is required not only for protein synthesis but also as the precursor for methylation reductions and for the biosynthesis of polyamines. Tumors have a much greater requirement for L-Met than normal tissues and become apoptotic when its availability is restricted. i.v. administration of bacterial (Pseudomonas) methionine-g-lyase is able to mediated near complete depletion of L-Met in serum and has been shown to drastically inhibit tumor growth of neuroblastomas, glioblastomas and prostate carcinomas in mouse xeongrafts. Furthermore strong synergistic effects with microtubule depolymerization agents have been reported. Unfortunately, in clinical trials the bacterial enzyme was shown to have very poor pharmacological properties (t 1/2 in serum only 2 hrs) and was found to be highly immunogenic in primates eliciting severe adverse responses that resulted in anaphylactic shock and death. While the human genome does not encode any methionine lyase enzymes, in preliminary studies we deployed protein engineering strategies to generate potentially non-immunogenic variants of the human enzyme cystathionine-g-lyase that: (a) exhibit high L-Met degradation activity in vitro and in vivo, (b) display a lower IC50 for neuroblastoma cell lines than their bacterial counterparts and (c) are about 10-fold more stable in mice. Here we will employ structure guided mutagenesis and directed evolution strategies to: 1. Engineer catalytically optimized human L-methioninases i.e. cystathionine-g-lyase enzymes with very high activity for L-Met degradation, even better stability in serum and high selectivity. 2. Develop optimized formats of the human L-methioninases for prolonged persistence in circulation by either site-specific PEGylation or by fusion to long intrinsically disordered polypeptide sequences (XTEN) and determine their pharmacokinetic and pharmacodynamic properties. 3. Evaluate the efficacy of these enzymes in the mouse xenograft model of human neuroblastoma tumors formed using clinical cell lines established either in diagnosis or relapse. The utility of these enzymes will be investigated both as monotherapy and in combination therapy with vincristine.